We considered randomised controlled trials (RCTs) or well designed quasi-experimental studies, controlled clinical trials (CCTs), controlled before and after studies (CBAs), and interrupted time series (ITS) analyses set in any type of healthcare organisation and investigating strategies to change organisational culture in order to improve healthcare performance. ITS analyses were eligible if they had a clearly defined point in time when the intervention occurred and three data collection points before and after the intervention to take into account secular trends and auto-correlation among measurements over time .
The two main outcomes of the review were: objective measures of professional performance such as prescription rates, the extent to which care is evidence based, quality of care; and objective measures of patient outcome such as mortality (standardised mortality ratio), condition-specific measures of outcome, quality of life, functional health status, and patients' satisfaction.
We also report other included outcomes such as: objective measures of organisational performance (such as wait times, inpatient hospital stay times, and staff turnover rates); measures of organisational culture; economic outcomes (such as efficiencies and changes in costs); and measures of health practitioners' knowledge, attitudes, satisfaction.
To identify studies eligible for this review we searched the following electronic databases for primary studies: The Cochrane Central Register of Controlled Trials (The Cochrane Library 2009, Issue 4), MEDLINE - Ovid (1950 to October Week 3 2009), EMBASE - Ovid (1980 to 2009 Week 41), CINAHL - EBSCO (1980 to October 2009), Sociological Abstracts - CSA (1952 to October 2009), Social Science Citation Index - Web of Knowledge (1970 to October 2009), Science Citation Index - Web of Knowledge (1970 to October 2009), Conference Proceedings - Web of Knowledge (1970 to October 2009), PsycINFO - Ovid (1806 to October Week 3 2009), Business and Management - OCLC FirstSearch (1995 to October 2009), EThOS (British Library), Index to Theses (1716 to October 2009), Intute, HMIC - Ovid (1979 to October 2009), SIGLE, Scopus (1823 to October 2009). Search strategies for primary studies incorporated the methodological component of the Cochrane Collaboration Effective Practice and Organisation of Care Review Group search strategy combined with selected index terms and free text terms. We translated the MEDLINE search strategy into the other databases using the appropriate controlled vocabulary as applicable. The full search strategies are presented in Additional File 1. We also searched the reference lists of all papers and relevant reviews identified, and we contacted experts in the field for advice on further potential studies. Finally, we searched the Database of Abstracts of Reviews of Effectiveness (DARE) for related reviews.
We downloaded all titles and abstracts retrieved by electronic searching to the reference management database EndNote, and removed duplicates. At least two review authors (from EP, GF, MPE) independently examined the remaining references. We excluded those studies that clearly do not meet the inclusion criteria and obtained copies of the full text of potentially relevant references. At least two review authors (from EP, GF, MES, MPE, NB) independently assessed the eligibility of retrieved papers and extracted the data using a specifically developed checklist. We used the same criteria as those outlined in the Cochrane Handbook for Systematic Reviews of Interventions to evaluate data  and we resolved any disagreement by discussion and the involvement of an arbitrator (MPE) as necessary.
The risk of bias of the eligible studies was evaluated independently by at least two reviewers using the following criteria: RCTs, CCTs, and CBAs were assessed for generation of allocation sequence, concealment of allocation, baseline outcome measurements, baseline characteristics, incomplete outcome data, blinding of outcome assessor, protection against contamination, selective outcome reporting, and other risks of bias. ITS designs were also assessed for the independence of the intervention from other changes, the pre-specified shape of the intervention, and whether the intervention was likely or unlikely to affect data collection. Data were reported in natural units. Where baseline results were available from RCTs, CCTs, and CBAs, we reported pre-intervention and post-intervention means or proportions for both study and control groups. We calculated the adjusted (for any baseline imbalance) absolute change from baseline reported as the adjusted risk difference (ARD) calculated as: (Intervention Follow-up - Intervention Baseline) - (Control Follow-up - Control Baseline).